Despite improvements in 1-year kidney allograft survival, chronic graft dysfunction (CGD) and subsequent late graft loss persists as major clinical problem. The objective of this proposal is to determine whether allelic variants of genes involved in regulation of immune response (via cytokines and chemokines), fibrosis, growth factors, vascular adhesion molecules and hypertension are associated with (1) CGD defined as persistent 25% increase in serum creatinine from a baseline established at 3 months post-transplantation and (2) persistent 25% decline in estimated glomerular filtration rate (eGFR), in a racially diverse transplant population. AIM 1 will study recipient candidate genes and AIM 2 will study the living donor genes. Our subaim #1 will compare the frequency of allelic variants that are associated with CGD and eGFR among African Americans and non-African American recipients. Our subaim #2 will study the interaction of recipient cytomegalovirus exposure and allelic variants of recipient immune response genes with CGDand eGFR. Understanding the genetic variants of potential determinants of CGD and eGFR may suggest better therapeutic approaches that extend the function of the kidney allografts. The research aims will be accomplished via a multicenter, prospective cohort of kidney transplant recipients enrolled in an ongoing NIH funded study studying kidney allograft biopsies during deterioration of kidney function. Secondary endpoints will be a composite of persistent 25% increase in serum creatinine and quantitative and semi-quantitative pathological findings on kidney biopsy. A test set of the first 1000 subjects will be genotyped using an Affymetrix custom SNP chip with approximately 3500 SNPs representing 1025 genes from 47 different biological pathways. Those SNPs with an association with the outcomes of interest in the Test cohort will be further genotyped in the subsequent Validation Cohort of 4000 transplant recipients for Aim 1 and for 2000 living donors for Aim2. The Validation cohort will also generate haplotypes of the selected candidate genes. A Subaim of Aim 2 will also explore the impact of multigene effects of living donor and recipient genotypes on CGD and a persistent 25% decline in eGFR. A proportional hazards model will be implemented to explore the relationship of the candidate gene polymorphisms with time to CGD or a 25% decline in eGFR. Haplotype analysis will also be conducted. Multi-gene effects will be explored using a variety of analytical techniques. These include statistical and probabilistic methods such as Bayesian classifiers and clustering as well as supervised machine learning methods such as decision tree induction and evolutionary computation-based learning classifier systems. By identifying genetic polymorphisms that impact CGD and eGFR, this study will help identify patients at risk and will help in the development of therapies targeting critical pathways.